DOCSLIB.ORG

Multilevel Logistic Regression Analysis Applied to Binary Contraceptive Prevalence Data

Total Page：16

File Type：pdf, Size：1020Kb

Download full-text PDF Read full-text
Multilevel Logistic Regression Analysis Applied to Binary Contraceptive Prevalence Data Journal of Data Science 9(2011), 93-110 Multilevel Logistic Regression Analysis Applied to Binary Contraceptive Prevalence Data Md. Hasinur Rahaman Khan and J. Ewart H. Shaw University of Warwick Abstract: In public health, demography and sociology, large-scale surveys often follow a hierarchical data structure as the surveys are based on mul- tistage strati¯ed cluster sampling. The appropriate approach to analyzing such survey data is therefore based on nested sources of variability which come from di®erent levels of the hierarchy. When the variance of the resid- ual errors is correlated between individual observations as a result of these nested structures, traditional logistic regression is inappropriate. We use the 2004 Bangladesh Demographic and Health Survey (BDHS) contraceptive bi- nary data which is a multistage strati¯ed cluster dataset. This dataset is used to exemplify all aspects of working with multilevel logistic regression models, including model conceptualization, model description, understand- ing of the structure of required multilevel data, estimation of the model via the statistical package MLwiN, comparison between di®erent estimations, and investigation of the selected determinants of contraceptive use. Key words: BDHS, CPR, cluster, division, MCMC, MLwiN, multilevel, MQL, PQL, TFR. 1. Introduction Bangladesh is the most densely populated country in the world. The country has currently a population about 150 million, with a corresponding population density of 939 per square kilometer and growth rate of 1.42% (M. Anwarul Iqbal, 2008). In the second half of the last century, the population grew extraordinarily rapidly, tripling during the period, whereas during the entire ¯rst half of the century the population increased by only 45%. Family planning was introduced in Bangladesh in the early 1950s. The policy to reduce fertility rates has been repeatedly rea±rmed by the Government of Bangladesh since liberation in 1971. During the mid 1970s, the contraceptive prevalence rate (CPR) was less than 10% and the total fertility rate (TFR) was more than 6 births per women (Islam and Islam, 1993). The subsequent last two rounds of the BDHS, in 1999¡2000 94 Md. H. R. Khan and J. E. H. Shaw and 2004, found CPRs of 54% and 58%, respectively, the TFRs for those years were 3.3 and 3.0 (NIPORT, 1999-2000, 2005). The association between estimated levels of contraceptive prevalence and the level of fertility is very close in the higher fertility countries like Bangladesh (Cur- tis and Diamond, 1995). As per the widely accepted correlation between CPR and TFR, a rise of 9% points in CPR has been seen to be accompanied by a fall of 0.64 in the TFR (Mauldin and Segal, 1988). Bangladesh still lacks such an estimate, which raises questions about the country's fertility and contracep- tive dynamics and prospects for future fertility decline. Use of contraception is the main contributor of fertility declining, as has been shown by many research workers (Cleland et al., 1994; R. Amin et al., 1994; Rani and Radheshyam, 2007). Fertility decline should continue if the wider use of contraception continues in all levels and groups of people in Bangladesh. It is critical for family planning workers to continue to meet the needs of existing family planning users, and also to address unmet need for family planning since individual tastes, interests, behaviours, etc. di®er from one unit to another within each level, owing to variability among various socioeconomic and geographical factors such as religion, culture, income, place of residence, education, occupation, mass media access, administrative and social facilities, and so on. That is why their e®orts and approaches do not seem to be equally e®ective, evenly served or acknowledged in some areas. As a result, the e®ectiveness of the program varies considerably. It is necessary to assess the within- and between-level variation, and to estimate the true e®ect of the above-mentioned factors on CPR, in order to implement more e®ective future family planning policies that target particular units at various levels of the hierarchy. This paper highlights the importance of multilevel analysis using logistic re- gression models for studying contraceptive prevalence in Bangladesh from the multistage clustered 2004 BDHS data. The paper aims to investigate the se- lected factors a®ecting the regulation of fertility through contraception in the context of multilevel modeling. It also aims to measure the influence of the com- bination of the selected factors on the current contraceptive practice of women in Bangladesh, and emphasis is given to exploring the true e®ect of the factors on the contraceptive prevalence taking into consideration the e®ect of the levels. The analysis is mainly carried out using MLwiN (Rasbash et al., 2004). 2. The Multilevel Model 2.1 Multilevel analysis for multistage clustered data In multilevel research, the structure of data in the population is hierarchical, and a sample from such a population can be viewed as a multistage sample. Multilevel Logistic Regression Analysis 95 Because of cost, time and e±ciency considerations, strati¯ed multistage samples are the norm for sociological and demographic surveys. For such samples the clustering of the data is, in the phase of data analysis and data reporting, a nuisance which should be taken into consideration. However, these samples, while e±cient for estimation of the descriptive population quantities, pose many challenges for model-based statistical inference. This clustering sampling scheme often introduces multilevel dependency or correlation among the observations that can have implications for model pa- rameter estimates. For multistage clustered samples, the dependence among observations often comes from several levels of the hierarchy. The problem of de- pendencies between individual observations also occurs in survey research, where the sample is not taken randomly but cluster sampling from geographical areas is used instead. In this case, the use of single-level statistical models is no longer valid and reasonable. Hence, in order to draw appropriate inferences and con- clusions from multistage strati¯ed clustered survey data we may require tricky and complicated modeling techniques like multilevel modeling, and very often the computation required for this is not straightforward and is not very time consuming as currently there is a number of software packages. The 2004 BDHS data set used for this study is based on multistage strati- ¯ed cluster sampling. The appropriate approach to analyzing contraceptive data from this survey is therefore based on nested sources of variability. Here the units at lower level (level-1) are individuals (ever-married women aged 10¡49) who are nested within units at higher level (clusters: level-2) and the clusters are again nested within units at the next higher level (divisions: level-3). Clusters are primary sampling units (PSU) de¯ned by the National Census of 1981, and correspond approximately to village in rural areas. All clusters are approximately of equal size in terms of area. On the other hand Divisions are administrative areas each of which consists of a number of sub-administrative areas called Zila. Due to this nested structure, the odds of women experiencing the outcome of interest are not independent, because women from the same cluster may share common exposure to community characteristics. The response variable in this study is \currently using contraception" which is binary and hence multilevel logistic regression model is a natural choice for modeling. Traditional logistic regression (which, in multilevel analysis terms, is single-level) requires the as- sumptions: (a) independence of the observations conditional on the explanatory variables and (b) uncorrelated residual errors. These assumptions are not always met when analyzing nested data. But the multilevel logistic regression analysis consider the variations due to hierarchy structure in the data. It allows the si- multaneous examination of the e®ects of group level (cluster and division) and individual level variables on individual level outcomes while accounting for the 96 Md. H. R. Khan and J. E. H. Shaw non-independence of observations within groups. Also this analysis allows the examination of both between group and within group variability as well as how group level and individual level variables are related to variability at both levels. 2.2 Parameter estimation The most common methods for estimating multilevel logistic models, used in this study, are based on likelihood. Among the methods, Marginal Quasi Like- lihood (MQL) (Goldstein, 1991; Goldstein and Rasbash, 1996) and Penalized Quasi Likelihood (PQL) (Laird, 1978; Breslow and Clayton, 1993) are the two prevailing approximation procedures. After applying these quasi likelihood meth- ods, the model is then estimated using iterative generalised least squares (IGLS) or reweighted IGLS (RIGLS) (Goldstein, 2003). Second-order PQL method has been used throughout the multi-level analyses since this method approximates well compared to the other PQL and MQL methods (Goldstein, 2003). Details of the PQL method are given below. Bayesian methods using Markov chain Monte Carlo (MCMC) have also been used for parameter estimation. Penalized quasi-likelihood The PQL estimation procedure is described here for two level logistic regres- sion models. Consider a level-1 outcome Yij taking on a value of 1 with conditional probability pij. Then the logit model or the generalized linear model is, · ¸ pij T T ln = ´ij = Xij γ + Zij uj 1 ¡ pij for level-1 unit i nested within level-2 unit j. At level 1, we assume Yij condition- ally distributed as Bernoulli, while the random e®ects vector uj is distributed as 2 2 N(0; σu) across the level-2 units. Let us consider the variance σu as T throughout this PQL estimation procedure. The PQL approach can be derived as a nonlinear regression model. In the case of binary outcomes with logit link, we start with the level-1 model Yij = pij + eij; (2.1) where E(eij) = 0 and V ar(eij) = pij(1 ¡ pij).
Load more

Recommended publications
  • A Primer for Analyzing Nested Data: Multilevel Modeling in SPSS Using
    REL 2015–046 The National Center for Education Evaluation and Regional Assistance (NCEE) conducts unbiased large-scale evaluations of education programs and practices supported by federal funds; provides research-based technical assistance to educators and policymakers; and supports the synthesis and the widespread dissemination of the results of research and evaluation throughout the United States. December 2014 This report was prepared for the Institute of Education Sciences (IES) under contract ED-IES-12-C-0009 by Regional Educational Laboratory Northeast & Islands administered by Education Development Center, Inc. (EDC). The content of the publication does not necessarily reflect the views or policies of IES or the U.S. Department of Education nor does mention of trade names, commercial products, or organizations imply endorsement by the U.S. Government. This REL report is in the public domain. While permission to reprint this publication is not necessary, it should be cited as: O’Dwyer, L. M., and Parker, C. E. (2014). A primer for analyzing nested data: multilevel mod­ eling in SPSS using an example from a REL study (REL 2015–046). Washington, DC: U.S. Department of Education, Institute of Education Sciences, National Center for Educa­ tion Evaluation and Regional Assistance, Regional Educational Laboratory Northeast & Islands. Retrieved from http://ies.ed.gov/ncee/edlabs. This report is available on the Regional Educational Laboratory website at http://ies.ed.gov/ ncee/edlabs. Summary Researchers often study how students’ academic outcomes are associated with the charac­ teristics of their classrooms, schools, and districts. They also study subgroups of students such as English language learner students and students in special education.
    [Show full text]
  • A Bayesian Multilevel Model for Time Series Applied to Learning in Experimental Auctions
    Linköpings universitet | Institutionen för datavetenskap Kandidatuppsats, 15 hp | Statistik Vårterminen 2016 | LIU-IDA/STAT-G--16/003—SE A Bayesian Multilevel Model for Time Series Applied to Learning in Experimental Auctions Torrin Danner Handledare: Bertil Wegmann Examinator: Linda Wänström Linköpings universitet SE-581 83 Linköping, Sweden 013-28 10 00, www.liu.se Abstract Establishing what variables affect learning rates in experimental auctions can be valuable in determining how competitive bidders in auctions learn. This study aims to be a foray into this field. The differences, both absolute and actual, between participant bids and optimal bids are evaluated in terms of the effects from a variety of variables such as age, sex, etc. An optimal bid in the context of an auction is the best bid a participant can place to win the auction without paying more than the value of the item, thus maximizing their revenues. This study focuses on how two opponent types, humans and computers, affect the rate at which participants learn to optimize their winnings. A Bayesian multilevel model for time series is used to model the learning rate of actual bids from participants in an experimental auction study. The variables examined at the first level were auction type, signal, round, interaction effects between auction type and signal and interaction effects between auction type and round. At a 90% credibility interval, the true value of the mean for the intercept and all slopes falls within an interval that also includes 0. Therefore, none of the variables are deemed to be likely to influence the model. The variables on the second level were age, IQ, sex and answers from a short quiz about how participants felt when they won or lost auctions.
    [Show full text]
  • Introduction to Multilevel Modeling
    1 INTRODUCTION TO MULTILEVEL MODELING OVERVIEW Multilevel modeling goes back over half a century to when social scientists becamedistribute attuned to the fact that what was true at the group level was not necessarily true at the individual level. The classic example was percentage literate and percentage African American. Using data aggregated to the state level for the 50 American states in the 1930s as units ofor observation, Robinson (1950) found that there was a strong correlation of race with illiteracy. When using individual- level data, however, the correlation was observed to be far lower. This difference arose from the fact that the Southern states, which had many African Americans, also had many illiterates of all races. Robinson described potential misinterpretations based on aggregated data, such as state-level data, as a type of “ecological fallacy” (see also Clancy, Berger, & Magliozzi, 2003). In the classic article by Davis, Spaeth, and Huson (1961),post, the same problem was put in terms of within-group versus between-group effects, corresponding to individual-level and group-level effects. A central function of multilevel modeling is to separate within-group individual effects from between-group aggregate effects. Multilevel modeling (MLM) is appropriate whenever there is clustering of the outcome variable by a categorical variable such that error is no longer independent as required by ordinary least squares (OLS) regression but rather error is correlated. Clustering of level 1 outcome scores within levels formed by a levelcopy, 2 clustering variable (e.g., employee ratings clustering by work unit) means that OLS estimates will be too high for some units and too low for others.
    [Show full text]
  • An Introduction to Bayesian Multilevel (Hierarchical) Modelling Using
    An Intro duction to Bayesian Multilevel Hierarchical Mo delling using MLwiN by William Browne and Harvey Goldstein Centre for Multilevel Mo delling Institute of Education London May Summary What Is Multilevel Mo delling Variance comp onents and Random slop es re gression mo dels Comparisons between frequentist and Bayesian approaches Mo del comparison in multilevel mo delling Binary resp onses multilevel logistic regres sion mo dels Comparisons between various MCMC meth ods Areas of current research Junior Scho ol Project JSP Dataset The JSP dataset Mortimore et al was a longitudinal study of around primary scho ol children from scho ols in the Inner Lon don Education Authority Sample of interest here consists of pupils from scho ols Main outcome of intake is a score out of in year in mathematics MATH Main predictor is a score in year again out of on a mathematics test MATH Other predictors are child gender and fathers o ccupation status manualnon manual Why Multilevel mo delling Could t a simple linear regression of MATH and MATH on the full dataset Could t separate regressions of MATH and MATH for each scho ol indep endently The rst approach ignores scho ol eects dif ferences whilst the second ignores similari ties between the scho ols Solution A multilevel or hierarchical or ran dom eects mo del treats both the students and scho ols as random variables and can be seen as a compromise between the two single level approaches The predicted regression lines for each scho ol pro duced by a multilevel mo del are a weighted average of the individual scho ol lines and the full dataset regression line A mo del with both random scho ol intercepts and slop es is called a random slop es regression mo del Random Slopes Regression (RSR) Model MLwiN (Rasbash et al.
    [Show full text]
  • 3 Diagnostic Checks for Multilevel Models 141 Eroscedasticity, I.E., Non-Constant Variances of the Random Effects
    3 Diagnostic Checks for Multilevel Models Tom A. B. Snijders1,2 and Johannes Berkhof3 1 University of Oxford 2 University of Groningen 3 VU University Medical Center, Amsterdam 3.1 Specification of the Two-Level Model This chapter focuses on diagnostics for the two-level Hierarchical Linear Model (HLM). This model, as defined in chapter 1, is given by y X β Z δ ǫ j = j + j j + j , j = 1, . , m, (3.1a) with ǫ ∅ Σ (θ) ∅ j , j (3.1b) δ ∼ N ∅ ∅ Ω(ξ) j and (ǫ , δ ) (ǫ , δ ) (3.1c) j j ⊥ ℓ ℓ for all j = ℓ. The lengths of the vectors yj, β, and δj, respectively, are nj, r, 6 and s. Like in all regression-type models, the explanatory variables X and Z are regarded as fixed variables, which can also be expressed by saying that the distributions of the random variables ǫ and δ are conditional on X and Z. The random variables ǫ and δ are also called the vectors of residuals at levels 1 and 2, respectively. The variables δ are also called random slopes. Level-two units are also called clusters. The standard and most frequently used specification of the covariance matrices is that level-one residuals are i.i.d., i.e., 2 Σj(θ)= σ Inj , (3.1d) where Inj is the nj-dimensional identity matrix; and that either all elements of the level-two covariance matrix Ω are free parameters (so one could identify Ω with ξ), or some of them are constrained to 0 and the others are free parameters.
    [Show full text]
  • Multilevel Analysis
    MULTILEVEL ANALYSIS Tom A. B. Snijders http://www.stats.ox.ac.uk/~snijders/mlbook.htm Department of Statistics University of Oxford 2012 Foreword This is a set of slides following Snijders & Bosker (2012). The page headings give the chapter numbers and the page numbers in the book. Literature: Tom Snijders & Roel Bosker, Multilevel Analysis: An Introduction to Basic and Applied Multilevel Analysis, 2nd edition. Sage, 2012. Chapters 1-2, 4-6, 8, 10, 13, 14, 17. There is an associated website http://www.stats.ox.ac.uk/~snijders/mlbook.htm containing data sets and scripts for various software packages. These slides are not self-contained, for understanding them it is necessary also to study the corresponding parts of the book! 2 2. Multilevel data and multilevel analysis 7 2. Multilevel data and multilevel analysis Multilevel Analysis using the hierarchical linear model : random coefficient regression analysis for data with several nested levels. Each level is (potentially) a source of unexplained variability. 3 2. Multilevel data and multilevel analysis 9 Some examples of units at the macro and micro level: macro-level micro-level schools teachers classes pupils neighborhoods families districts voters firms departments departments employees families children litters animals doctors patients interviewers respondents judges suspects subjects measurements respondents = egos alters 4 2. Multilevel data and multilevel analysis 11{12 Multilevel analysis is a suitable approach to take into account the social contexts as well as the individual respondents or subjects. The hierarchical linear model is a type of regression analysis for multilevel data where the dependent variable is at the lowest level.
    [Show full text]
  • Multilevel Linear Models: the Basics
    CHAPTER 12 Multilevel linear models: the basics Multilevel modeling can be thought of in two equivalent ways: We can think of a generalization of linear regression, where intercepts, and possi- • bly slopes, are allowed to vary by group. For example, starting with a regression model with one predictor, yi = α + βxi + #i,wecangeneralizetothevarying- intercept model, yi = αj[i] + βxi + #i,andthevarying-intercept,varying-slope model, yi = αj[i] + βj[i]xi + #i (see Figure 11.1 on page 238). Equivalently, we can think of multilevel modeling as a regression that includes a • categorical input variable representing group membership. From this perspective, the group index is a factor with J levels, corresponding to J predictors in the regression model (or 2J if they are interacted with a predictor x in a varying- intercept, varying-slope model; or 3J if they are interacted with two predictors X(1),X(2);andsoforth). In either case, J 1linearpredictorsareaddedtothemodel(or,toputitanother way, the constant− term in the regression is replaced by J separate intercept terms). The crucial multilevel modeling step is that these J coefficients are then themselves given a model (most simply, a common distribution for the J parameters αj or, more generally, a regression model for the αj’s given group-level predictors). The group-level model is estimated simultaneously with the data-level regression of y. This chapter introduces multilevel linear regression step by step. We begin in Section 12.2 by characterizing multilevel modeling as a compromise between two extremes: complete pooling,inwhichthegroupindicatorsarenotincludedinthe model, and no pooling, in which separate models are fit within each group.
    [Show full text]
  • Introduction to Multilevel Models for Longitudinal and Repeated Measures Data
    Introduction to Multilevel Models for Longitudinal and Repeated Measures Data • Today’s Class: Features of longitudinal data Features of longitudinal models What can MLM do for you? What to expect in this course (and the next course) CLDP 944: Lecture 1 1 What is CL(D)P 944 about? • “Longitudinal” data Same individual units of analysis measured at different occasions (which can range from milliseconds to decades) • “Repeated measures” data (if time permits) Same individual units of analysis measured via different items, using different stimuli, or under different conditions • Both of these fall under a more general category of “multivariate” data of varying complexity The link between them is the use of random effects to describe covariance of outcomes from the same unit CLDP 944: Lecture 1 2 Data Requirements for Our Models • A useful outcome variable: Has an interval scale* . A one-unit difference means the same thing across all scale points . In subscales, each contributing item has an equivalent scale . *Other kinds of outcomes will be analyzed using generalized multilevel models instead, but estimation will be more challenging Has scores with the same meaning over observations . Includes meaning of construct . Includes how items relate to the scale . Implies measurement invariance • FANCY MODELS CANNOT SAVE BADLY MEASURED VARIABLES OR CONFOUNDED RESEARCH DESIGNS. CLDP 944: Lecture 1 3 Requirements for Longitudinal Data • Multiple OUTCOMES from same sampling unit (person) 2 is the minimum, but just 2 can lead to problems: . Only 1 kind of change is observable (1 difference) . Can’t distinguish “real” individual differences in change from error . Repeated measures ANOVA is just fine for 2 observations – Necessary assumption of “sphericity” is satisfied with only 2 observations even if compound symmetry doesn’t hold More data is better (with diminishing returns) .
    [Show full text]
  • Multilevel Models for Repeated Binary Outcomes: Attitudes and Vote Over the Electoral Cycle
    Multilevel models for repeated binary outcomes: attitudes and vote over the electoral cycle by Min Yang,* Anthony Heath** and Harvey Goldstein* * Institute of Education, University of London; ** Nuffield College, Oxford ABSTRACT Models for fitting longitudinal binary responses are explored using a panel study of voting intentions. A standard multilevel repeated measures logistic model is shown to be inadequate due to the presence of a substantial proportion of respondents who maintain a constant response over time. A multivariate binary response model is shown to be a better fit to the data. SOME KEYWORDS Longitudinal binary data, multivariate multilevel model, multilevel, political attitudes, voting. ACKNOWLEDGEMENTS This work was carried out as part of the Multilevel Models Project funded by the Economic and Social Research Council (UK) under the programme for the Analysis of Large and Complex Datasets. We are very grateful to referees for comments on an earlier draft. 1. INTRODUCTION The electoral cycle has become an established feature of voting behaviour, both in Britain and in other European countries. After an initial ‘honeymoon’ between a new government and the electorate, disillusion often sets in and government popularity - whether measured by opinion polls, by-elections or midterm elections such as the European and local elections - tends to decline. In most cases, there is then some recovery in the government’s standing in the run-up to the next general election (Miller, Tagg and Britton, 1986; Miller and Mackie, 1973; Reif, 1984; Stray and Silver, 1983). During the 1987-92 British parliament, for example, the Conservative government lost seven by-elections but subsequently won all of them back at the 1992 general election.
    [Show full text]
  • User-Friendly Bayesian Regression Modeling: a Tutorial with Rstanarm and Shinystan
    ¦ 2018 Vol. 14 no. 2 User-friendly BaYESIAN REGRESSION modeling: A TUTORIAL WITH rstanarm AND shinystan Chelsea Muth a, B, Zita OrAVECZ A & Jonah Gabry B A Pennsylvania State University B Columbia University AbstrACT ACTING Editor This TUTORIAL PROVIDES A PRAGMATIC INTRODUCTION TO specifying, ESTIMATING AND interpret- De- ING single-level AND HIERARCHICAL LINEAR REGRESSION MODELS IN THE BaYESIAN FRamework. WE START BY NIS Cousineau (Uni- versité D’Ottawa) SUMMARIZING WHY ONE SHOULD CONSIDER THE BaYESIAN APPROACH TO THE MOST COMMON FORMS OF regres- Reviewers sion. Next WE INTRODUCE THE R PACKAGE rstanarm FOR BaYESIAN APPLIED REGRESSION modeling. An OVERVIEW OF rstanarm FUNDAMENTALS ACCOMPANIES step-by-step GUIDANCE FOR fiTTING A single-level One ANONYNOUS re- viewer. REGRESSION MODEL WITH THE stan_glm function, AND fiTTING HIERARCHICAL REGRESSION MODELS WITH THE stan_lmer function, ILLUSTRATED WITH DATA FROM AN EXPERIENCE SAMPLING STUDY ON CHANGES IN af- FECTIVE states. ExplorATION OF THE RESULTS IS FACILITATED BY THE INTUITIVE AND user-friendly shinystan package. Data AND SCRIPTS ARE AVAILABLE ON THE Open Science FrAMEWORK PAGE OF THE project. FOR READERS UNFAMILIAR WITH R, THIS TUTORIAL IS self-contained TO ENABLE ALL RESEARCHERS WHO APPLY regres- SION TECHNIQUES TO TRY THESE METHODS WITH THEIR OWN data. Regression MODELING WITH THE FUNCTIONS IN THE rstanarm PACKAGE WILL BE A STRAIGHTFORWARD TRANSITION FOR RESEARCHERS FAMILIAR WITH THEIR FREQUENTIST counterparts, lm (or glm) AND lmer. KEYWORDS TOOLS BaYESIAN modeling, regression, HIERARCHICAL LINEAR model. Stan, R, rstanarm. B [email protected] CM: n/a; ZO: 0000-0002-9070-3329; JG: n/a 10.20982/tqmp.14.2.p099 Introduction research.
    [Show full text]
  • Fundamentals of Hierarchical Linear and Multilevel Modeling 1 G
    Fundamentals of Hierarchical Linear and Multilevel Modeling 1 G. David Garson INTRODUCTION Hierarchical linear models and multilevel models are variant terms for what are broadly called linear mixed models (LMM). These models handle data where observations are not independent, correctly modeling correlated error. Uncorrelated error is an important but often violated assumption of statistical procedures in the general linear model family, which includes analysis of variance, correlation, regression, and factor analysis. Violations occur when error terms are not independent but instead cluster by one or more grouping variables. For instance, predicted student test scores and errors in predicting them may cluster by classroom, school, and municipality. When clustering occurs due to a grouping factor (this is the rule, not the exception), then the standard errors computed for prediction parameters will be wrong (ex., wrong b coefficients in regression). Moreover, as is shown in the application in Chapter 6 in this volume, effects of predictor variables may be misinterpreted, not only in magnitude but even in direction. Linear mixed modeling, including hierarchical linear modeling, can lead to substantially different conclusions compared to conventional regression analysis. Raudenbush and Bryk (2002), citing their 1988 research on the increase over time of math scores among students in Grades 1 through 3, wrote that with hierarchical linear modeling, The results were startling—83% of the variance in growth rates was between schools. In contrast, only about 14% of the variance in initial status was between schools, which is consistent with results typically encountered in cross-sectional studies of school effects. This analysis identified substantial differences among schools that conventional models would not have detected because such analyses do not allow for the partitioning of learning-rate variance into within- and between-school components.
    [Show full text]
  • Actuarial Applications of Hierarchical Modeling
    Actuarial Applications of Hierarchical Modeling CAS RPM Seminar Jim Guszcza Chicago Deloitte Consulting LLP March, 2010 Antitrust Notice The Casualty Actuarial Society is committed to adhering strictly to the letter and spirit of the antitrust laws. Seminars conducted under the auspices of the CAS are designed solely to provide a forum for the expression of various points of view on topics described in the programs or agendas for such meetings . Under no circumstances shall CAS seminars be used as a means for competing companies or firms to reach any understanding – expressed or implied – that restricts competition or in any way impairs the ability of members to exercise independent business judgment regarding matters affecting competition. It is the responsibility of all seminar participants to be aware of antitrust regulations, to prevent any written or verbal discussions that appear to violate these laws, and to adhere in every respect to the CAS antitrust compliance policy. Copyright © 2009 Deloitte Development LLC. All rights reserved. 1 Topics Hierarchical Modeling Theory Sample Hierarchical Model Hierarchical Models and Credibilityyy Theory Case Study: Poisson Regression Copyright © 2009 Deloitte Development LLC. All rights reserved. 2 Hierarchical Model Theory Hierarchical Model Theory What is Hierarchical Modeling? • Hierarchical modeling is used when one’s data is grouped in some important way. • Claim experience by state or territory • Workers Comp claim experience by class code • Income by profession • Claim severity by injury type • Churn rate by agency • Multiple years of loss experience by policyholder. •… • Often grouped data is modeled either by: • Pooling the data and introducing dummy variables to reflect the groups • Building separate models by group • Hierarchical modeling offers a “third way”.
    [Show full text]